Locked lesson.

Upgrade

About this lesson

A Lean Six Sigma Black Belt will often chair the stage gate review meetings for Lean Six Sigma projects. In those meetings, the Black Belt needs to ensure the work of the phase was done and the tools were used effectively. This lesson reviews the normal deliverables due at the Measure stage gate review. It also includes hints and tips for identifying problems to be avoided during that phase.

Quick reference

Measure Stage Deliverables

The Measure Stage is the second stage of a Lean Six Sigma project.  The deliverables from this stage are the problem statements, data, and metrics used for the analysis that occurs in future stages of the project.

When to use

The Measure Stage deliverables should be reviewed and approved at the Measure Stage Gate Review meeting.

Instructions

Throughout the Measure Stage, the Lean Six Sigma team creates the deliverables.  The deliverables are the data that is used to clarify the problem statement and the process map or value stream map with data that demonstrates current performance.  Normally, the problem statement and process map/value stream map are reviewed and approved during the Measure Stage Gate Review.   These maps and the problem statement provide a clear focus for the analysis that will occur in later stages.  While data is usually collected for each step in the process map/value stream map; the data collection is focused in the area of the perceived defect or problem.     The final result of this stage is a data-based problem statement that describes in measurable terms the magnitude and characteristics (when and where) of the defect.

The focus of this stage is collecting reliable and relevant data.  To ensure the data is reliable, a measurement systems analysis is often needed.  There is an entire GoSkills course addressing this topic.  The bottom line is that the measurement system error (and there is always a measurement system error) should be very small with respect to the measurement variation.  This data will serve as the baseline process performance that is used for analysis of the process and to demonstrate that the improvement that is implemented has made a significant difference in the process performance.

The Lean Six Sigma equation is Y=f(x).  In this stage, the “x” terms are defined and data is collected for each term that establishes the range of the term.  Be certain when collecting data from multiple sources that you are using consistent units.

A major challenge for effective analysis in later stages will occur if the problem statement is not clear or the data is inaccurate.  When that occurs, the analysis done in later stages is not able to establish root causes.  As a Black Belt, you will need to review the process for data collection and you will want to ensure the problem statement is specific without assuming a solution. Questions you often must work through are:

  • Is this the correct Problem Statement and  Process Map/Value Stream Map? Have you walked the process and does the data support this problem statement?
  • What data sources and data sets were used? Are you confident that the measurements are accurate?
  • What is the real baseline process performance? What is the magnitude of the defect?

Hints & tips

  • Use historical data where it is available.  This data is usually faster to obtain and you don’t need to worry about the Hawthorne Effect.
  • Minimize the Hawthorne Effect as much as possible.  This effect usually results in different performance by operators (normally better) because they know someone is watching and measuring what they are doing.   Try not to introduce any new data collection steps or actions in the process.
  • If your data collection is manual, do a measurement systems analysis, especially in and around the defect zone of the project, to ensure that you have valid data.  I had a project one time that had already failed several improvement attempts.  When I started to work with the team, I insisted on an MSA for the primary inspection measure.  We found that the measurement error was nearly three times larger than the allowed tolerance and more than five times larger than the normal product variation.  The measurement system was the problem in this case – not the process.
  • Know the metadata for your data sets, especially if the data set was not collected by your team.  You need to ensure you know what was measured, how it was measured, and the measurement system or units.  It is as important to know what was not measured as it is to know what was measured.
  • Document your baseline performance so that you can demonstrate your solution has made a difference.
  • Do a thorough walk-through of your process map/value stream map.  There are often overlooked steps that do not make it onto the map.  It is also a good idea to do multiple walk-throughs to minimize the Hawthorne Effect.
  • Don’t jump to a conclusion from the first piece of data.  Collect all your data and save the analysis for the next stage.  Otherwise, you are likely to jump at the first root cause you see and may miss other important ones.
Login to download
  • 00:04 Hi, I'm Ray Sheen.
  • 00:06 I want to summarize this section by discussing the measure stage deliverables
  • 00:10 and some of the challenges that you may encounter while those are being developed.
  • 00:15 Let's start by looking at the gate review and
  • 00:17 what is required to get through that meeting.
  • 00:20 The focus of the gate review is to collect reliable and
  • 00:23 relevant data that establishes the specific nature of the problem.
  • 00:27 This means that you need an accurate process map so that we can clarify
  • 00:31 where the problem exists, and the data demonstrates the magnitude of the problem.
  • 00:37 This means that our critical takeaways are the process map per value stream map and
  • 00:41 a clearly defined problem statement.
  • 00:44 With those, we can say that we have specified, this is the problem.
  • 00:49 The decisions that occur at this gate is approval of the problem statement.
  • 00:54 The statement should not assume a solution, but
  • 00:57 should clearly state the defect with the measurable criteria.
  • 01:02 And of course, this points to the biggest risk or challenge in this phase.
  • 01:06 That is, a problem statement that is either too vague or ambiguous on one hand,
  • 01:10 or so specific that it assumes the root cause in the problem statement.
  • 01:15 The problem statement should not include the word because.
  • 01:20 However, It should have a measurable defect that directly relates to
  • 01:24 the customer CTQ from the define stage.
  • 01:27 Another major challenge in this stage is the collection or aggregation of data that
  • 01:32 is not accurate or that does not reflect the actual process performance.
  • 01:36 The data must be relevant.
  • 01:38 An MSA should show that it can be trusted and
  • 01:41 you have minimized the Hawthorne effect.
  • 01:44 As you prepare for the gate review and sometimes during the gate review,
  • 01:48 you can expect several questions to come up again and again.
  • 01:52 Challenge the team to be certain that there is a good answer to each of
  • 01:56 these questions.
  • 01:57 Is this the correct problem statement and process map or value stream map?
  • 02:01 Have you walked the process?
  • 02:03 Is the problem statement based upon the data collected?
  • 02:07 What data sources and data sets were used?
  • 02:10 Did you rely on historical data?
  • 02:12 Can you trust it?
  • 02:13 Did you collect new data?
  • 02:15 Is the measurement system effective?
  • 02:17 How do you know?
  • 02:19 What is the real baseline process performance?
  • 02:22 Is the situation as bad as we thought, worse or better?
  • 02:26 And what is the current process capability or sigma level?
  • 02:31 Let's look at a quick process map of this phase.
  • 02:34 This phase is usually pretty straightforward.
  • 02:36 Decide what to measure and then measure it.
  • 02:39 The one loop in the process is associated with the measurement systems analysis.
  • 02:44 We have a short course that addresses the concerns with the measurement system that
  • 02:48 may require a team to loop back through that portion of this phases process.
  • 02:54 Of course, you don't want to hold a gate review if you're not ready.
  • 02:57 So let's look at the typical criteria for what is needed for
  • 03:01 successful measure stage gate review.
  • 03:03 The primary decision is the approval of the problem statement.
  • 03:08 So a key to success is to have a complete and factual based problem statement.
  • 03:13 You will also normally present a process map or value stream map to illustrate
  • 03:18 the location in the process of the defect or defects.
  • 03:22 This also will help to illustrate current process performance.
  • 03:26 Be prepared to explain your data sources, both the existing ones you used and
  • 03:30 new ones that you created.
  • 03:33 You don't need to show the raw data, but summarized data that
  • 03:36 illustrates the process baseline performance is a good thing to have.
  • 03:41 And of course, you can expect to get questions about the accuracy and validity
  • 03:46 of the data, especially if it appears to put one department in a bad light.
  • 03:50 Have your MSA results ready to discuss, if needed.
  • 03:55 I have found that the stakeholders often want to talk about the baseline
  • 03:58 performance.
  • 03:59 Many times they've only had a vague sense that something's wrong in this area.
  • 04:03 That is why they wanted to Lean Six Sigma project.
  • 04:07 Now, you have real data that shows real problems.
  • 04:10 They often want to dig in and understand the baseline performance with respect to
  • 04:14 things like cycle time, defect rates, throughput and process capability.
  • 04:19 If your data shows a time dependency to the problem, highlight this also.
  • 04:26 Let's finish this by providing a few tips and hints to help you get through this
  • 04:30 stage and have a successful stage review meeting.
  • 04:34 Leverage the historical data.
  • 04:36 It can be a goldmine of information.
  • 04:38 It may not have everything you need, but often it will have most of what is needed.
  • 04:44 Also be aware that when collecting data,
  • 04:47 you're introducing a possible Hawthorne effect.
  • 04:50 People perform differently when they know they're being watched.
  • 04:54 So try to keep that intrusion to a minimum.
  • 04:57 Since the problem statement is based upon the defect data collected,
  • 05:01 make sure the data is accurate.
  • 05:03 If there's any place that you may need an MSA or Gage R&R, this is the spot.
  • 05:10 If you're using a data set that was generated historically or
  • 05:13 from another project, be sure you understand what each data element means.
  • 05:18 This is referred to as the meta-data for the dataset.
  • 05:23 Establish the baseline performance, not only in the area of the defect, but for
  • 05:28 the entire process.
  • 05:29 When the solution is implemented, you'll need to be able to show that it
  • 05:33 has improved the defect area, while not degrading performance in other areas.
  • 05:39 Do a walk-through of the process.
  • 05:41 In fact, do several walk-throughs so as to minimize the Hawthorne effect.
  • 05:46 You want to be able to say with confidence that the process map is accurate.
  • 05:51 And don't rush to analyze.
  • 05:53 Get all your data.
  • 05:55 Jumping too soon to a conclusion will likely
  • 05:57 get you to an incomplete picture of the root cause or causes of the defects.
  • 06:04 As a Lean Six Sigma Black Belt, it's your job to make sure that the deliverables for
  • 06:08 the measure stage are completed, both for the projects that you're leading and for
  • 06:12 the projects that you're coaching.
  • 06:14 You need to ask the hard questions of the team before you go to the gate review.

Lesson notes are only available for subscribers.

PMI, PMP, CAPM and PMBOK are registered marks of the Project Management Institute, Inc.

Gift this course